Ataxia-telangiectasia (A-T) is an autosomal recessive multisystem disease in which homozygotes have an exceptionally high incidence of cancer. The hypothesis that A-T heterozygotes also have an excess cancer risk is important because, according to the Hardy-Weinberg principle, such heterozygotes are relatively common (0.5 - 5%) in the general population. Retrospective analysis of cancer incidence in an ongoing study of 114 white A-T and 15 black A-T families in the United States supports the hypothesis that the A-T heterozygote is predisposed to cancer. While the analysis is not yet complete, there is strong evidence that the A-T heterozygote is predisposed to pancreas and breast cancer. If these observations are confirmed by the proposed prospective and case-control components of this project, the A-T gene would be the first gene identified to be associated with these cancers in the general population. Using presently available techniques, the proposed prospective observation and analysis of cancer incidence of an augmented sample of A-T families will provide the most contemporary and least biased estimate of the excess mortality and cancer risk of the A-T heterozygote. Further, the extension of this prospective study will make it possible to critically test the hypotheses developed or supported by these retrospective and prospective family studies through a case-control study in which laboratory tests will determine directly which family members carry the A-T gene; such tests are being refined now and early results are promising. Since the living family members with cancer who are crucial to this case-control design are at high risk of being lost through death, it is planned that cell cultures from such persons be obtained as soon as possible, while tests for the A-T heterozygote are being further developed and validated. Because of our extensive casefinding, a great deal of systematic information about the A-T homozygote will be collected and be of immediate practical use to clinicians caring for these patients.